3 results for “flow matching”
Apple researchers introduced probe guidance, which steers diffusion language models using frozen internal states rather than a second pass. It reports state-of-the-art generation quality and stronger multiple-choice QA.
MiniMax released Music 3.0, an open-weight model that generates complete songs up to five minutes with structured vocals and evolving arrangements. The model combines an 8B Global LLM with a continuous Flow Matching synthesis pipeline for 32 kHz stereo output.
Black Forest Labs has introduced FLUX 3, a new multimodal foundation model that jointly learns from images, video, and audio within a single unified. The unified multimodal approach addresses a fundamental limitation of modality-specific.